I am currently experimenting with the LoRa learning settings for Flux, so here are the experimental models that will be updated in the future.
Personal Ami style
PRSNL.safetensors